Re: naming convention question

[Manoj Srivastava <srivasta@datasync.com>]

Hi,
>>"Dave" == Dave Restall <dave.restall@mpn.cp.philips.com> writes:

Dave> This is essentially the point.  As an end user I find the names
Dave> confusing.  It would help me as an end user if instead of
Dave> posters saying :-

Dave> "I'm using package XYZ from bilbo(or whatever the codename of
Dave> the release is) and problem such-and-such arises"

Dave> they would say :-

Dave> "I'm using package XYZ from the latest stable release etc.."


	I see your point. You are correct in the sense that just
 specifying a codename does not adequately specify which version of a
 package is being talked about (espescially when one is talking about
 the development versions, since the version numbers of packages are
 in a state of flux there). I generally say I am using version XYZ of
 package MMM.

	For the most part, when talking about a package, the version
 number of the package is important. When talking about
 releases, one uses numbers. Like Debian 1.3.1 rather than whatever it
 is called now (bo-updates??)

Manoj> Yup. Though of course, the general public knows stable,
Manoj> unstable, and Debian 1.1.3, for example. You don't *have* to
Manoj> know the names.

Dave> Exactly, therefore why have names ?  Forget I asked that :-), I
Dave> can understand the reasons for names and the like, I just have
Dave> problems dealing with them and trying to keep track of them,
Dave> especially when a name can relate to several different release
Dave> numbers.  Surely the numbers are more important than the names
Dave> and with the packages being dynamic even in 'stable' releases,
Dave> then numbers assume even more importance.

	You are correct. Stable releases should be identified by the
 numbers. 

	The names are used while the release is under development (we
 have been burned once by an over-eager CD retailer, we really don't
 need that kind of PR), and an artifact of mirroring technology that
 make renaming directories at release a bad idea for the mirrors.

	Also, it does give us some freedom in version numbers of the
 distribution (we can slip in a version 1.4 next, or go to 2.0), which
 of course means little to the end users.

Dave> I also run 1.2 something or other which I've built via ftp (Yes
Dave> I know it take ages but I got fed up with waiting for the CD
Dave> which when it did arrive had 1.1.16 on it instead of 1.2 - but
Dave> that's another story), every fortnight or so I run dselect and
Dave> download upgraded packages. This means that I have the 'stable'
Dave> release but my 'stable' release of is possibly quite different
Dave> from somebody elses 'stable' release of 1.2.  What I'm aiming at
Dave> here is that names are quite meaningless and numbers aren't that
Dave> much better, however you can uniquely identify a system using
Dave> numbers, you can't using names UNLESS you release a new name for
Dave> the release whenever some package in it changes.

	Quite so. Unfortunately, I think that the codenames offer
 enough technical benefits to the Debian support organizations that
 the (IMHO minor) irritation caused to the users is deemed
 acceptable. 

	I, on behalf of the Debian developers, apologize for the
 inconvenience caused.

	manoj

-- 
 "The only way for a reporter to look at a politician is down."
 H.L. Mencken
Manoj Srivastava               <url:mailto:srivasta@acm.org>
Mobile, Alabama USA            <url:http://www.datasync.com/%7Esrivasta/>


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-user-request@lists.debian.org . 
Trouble?  e-mail to templin@bucknell.edu .